While dietary and physical activity habits are known to relate to cardiovascular risk, little data are available on factors that influence the development of such behaviors early in life. In 1987, 106 families with 3 to 50 year old children were recruited to take part in the Framingham Children's Study. The children and their parents are direct descendents (4th and 3rd generation, respectively) of participants in the Framingham Heart Study. The families are currently undergoing Cycle 3 examinations of their eating behaviors, nutrient intake, physical activity habits, and a variety of potential determinants of such behaviors. During the initial grant period, the feasibility of collecting relevant data has been demonstrated. Maintenance of the cohort and acceptance of the monitoring procedures have been excellent, due partly to the fact that the families consider themselves a part of the Framingham Heart Study and take pride in long-term compliance with study procedures. The present application is for the continued follow-up of these families, especially to determine how peer, school, and community factors influence the children's behaviors. Data collected over the first three cycles (covering ages 3-7 years for the children, their "preschool years") are being used to characterized baseline eating and exercise habits. The baseline characteristics show familial aggregation, reflecting both genetic and family influences. The extension of the project will permit continued follow up of the children as they go through elementary school (ages 7-12), when influences on their behaviors from peers, school, and community are expected to become greater. The stability or lack of stability of the baseline eating and exercise behaviors over the elementary school years will be monitored, and environmental factors related to change will be sought. Information on the changing sources of food, peer influences, and availability and participation in school and community-based athletic opportunities will be collected. In addition, data on eating and exercise patterns of parents and information available from the Framingham Heart Study on the grandparents and great-grandparents of the children will be considered as potential determinants of the children's behaviors. Data from this study should identify early determinants of eating and exercise behaviors that relate to cardiovascular risk, and could lead to approaches for the prevention of cardiovascular disease.